Sepsis: Neonates require high suspicion, quick action
Combating neonatal sepsis requires a high index of suspicion and timely appropriate treatment. This approach can help minimize the burden of neonatal sepsis on families, physicians, and the healthcare system alike.
Despite the advent of intrapartum antibiotic prophylaxis (IAP) and ongoing attempts to identify more accurate diagnostic tools, neonatal sepsis or bacteremia remains a common and potentially deadly occurrence, particularly in very-low-birth-weight (VLBW, <1500 g) babies. Sepsis causes up to 26% of all VLBW infant deaths,1,2 and surviving VLBW infants face higher risks of problems including bronchopulmonary dysplasia, neurodevelopment impairment, and prolonged hospital stays. Because sepsis can have devastating consequences for full-term, normal-body-weight infants as well, early diagnosis and treatment remain crucial.
Incidence
Neonatal sepsis is a blood infection occurring during the neonatal period, whether one defines it as 1 month3 or 3 months.4 Overall, the condition affects approximately 1 to 5 per 1000 live births, with a lower incidence (1 to 2 per 1000) in term infants (those born after at least 37 weeks’ gestation).5 Conversely, in VLBW infants, a population-based cohort study showed that sepsis rates are 7 times higher than in infants with birth weight ≥ 2500 g.6 Somewhat similarly, less developed regions report overall sepsis rates much higher than average; eg, up to 38 per 1000 in India.7
As for mortality, a long-term US single-institution study showed that the sepsis-related overall mortality rate fell from 87% in 1928 to 3% in 2003.8 By contrast, 6 selected facilities in Bangladesh, Bolivia, Ghana, India, Pakistan, and South Africa recently reported a combined sepsis mortality rate of 7%.9
Etiology
Physicians classify sepsis according to its age of onset. Responsible pathogens tend to vary with the timing of disease onset: early-onset sepsis/EOS (first 3 days of life) or late-onset sepsis/LOS (after day 4).2
EOS
Most commonly, EOS stems from vertical transmission of group B Streptococcus (GBS) from mother to infant. According to figures from the Centers for Disease Control and Prevention (CDC), the rate of early-onset GBS invasive infection has declined from 0.6 per 1000 births in 2000 to 0.26 per 1000 in 2011.10,11 Early-onset sepsis because of GBS has a fatality rate of 5% to 20%.12
In recent years, Escherichia coli has emerged as the major pathogen responsible for sepsis in preterm infants and the second most common cause in full-term babies.13 Together, E coli and GBS account for 70% of EOS cases.8
The only intervention proven to reduce the incidence of GBS sepsis is giving expectant mothers intrapartum IV antibiotics—preferably penicillin—for at least 4 hours before delivery. Such a regimen is approximately 90% effective in preventing early-onset GBS disease.14 Alternate antibiotics can include the following:
- Ampicillin;
- Cefazolin (in mothers with mild penicillin allergy);
- Clindamycin (for women with life-threatening penicillin allergy whose GBS isolates are proven susceptible); and
- Vancomycin (if the clindamycin sensitivity is unknown or the GBS isolate is resistant).
Also regarding prevention, a multistate retrospective cohort study showed that universally screening women at 35 to 37 weeks of pregnancy and providing IAP for those with rectovaginal GBS colonization—an approach recommended by the CDC, the American Academy of Pediatrics (AAP), and the American College of Obstetricians and Gynecologists—resulted in notably lower GBS risk than the previous strategy of screening only women with risk factors for GBS disease.15
LOS
Late-onset sepsis occurs between day 4 and 120,2 usually resulting from postnatal horizontal transmission of pathogens.8 Preterm VLBW infants face a particularly high risk of LOS partly because of their immature immune systems and requirements for prolonged hospitalization, which often exposes them to mechanical ventilation, catheters, and other invasive procedures.
While gram-negative organisms cause most EOS, a 6215-infant National Institute of Child Health and Human Development (NICHD) study showed that 70% of LOS in VLBW babies stemmed from gram-positive organisms; coagulase-negative staphylococci accounted for 48% of first-episode infections in this cohort.6
Signs and symptoms
Because the clinical manifestations of neonatal sepsis are nonspecific, often subtle, and variable based on the causative agent (Table 1),4,12,16 its signs and symptoms can be difficult to distinguish from other causes of neonatal distress.
Evaluation and diagnosis
The nonspecific, often subtle signs and symptoms of neonatal sepsis require physicians to perform laboratory tests in any infant with identifiable risk factors, suspicious physical findings, or any deviations from the usual pattern of activity or feeding. In other words, perhaps the most important tip-off may be that the septic newborn just “doesn’t look right.”17
A routine newborn assessment includes a comprehensive physical examination, as well as a review of the pregnancy, labor, and delivery, with an eye toward risk factors for sepsis18-20:
- Intrapartum maternal temperature ≥100.4°F (38°C);
- Overt infection such as urinary tract infection (UTI);
- Maternal GBS colonization or history of previous infection;
- Maternal history of multiple obstetric procedures (including cervical sutures);
- Chorioamnionitis;
- Membrane rupture ≥18 hours;
- Delivery at <37 weeks’ gestation;
- 5-minute Apgar score ≤6;
- Evidence of fetal distress.
Among protective factors, physicians also should assess the adequacy of maternal IAP. However, the fact that a mother underwent IAP does not necessarily rule out sepsis.
In diagnosing neonatal sepsis, isolating bacteria from blood via conventional culturing is considered the gold standard. However, no definitive diagnostic tests for neonatal sepsis exist.21
Conventional culturing takes 24 to 48 hours for results, and requires 6 mL of blood for optimal results, which is not feasible with newborns. Blood cultures moreover may deliver false negative results in about 10% of cases.
To overcome such limitations, physicians use a constellation of clinical observations and laboratory criteria to identify newborns at significant risk for sepsis, so that empiric antibiotic treatment may begin while clinicians await blood culture results.
Because the nonspecific clinical signs of early sepsis also may mark other neonatal diseases such as respiratory distress syndrome (RDS), metabolic disorders, and intracranial hemorrhage, physicians also must rule out these entities while treating for neonatal sepsis.
Symptomatic infants
For any newborn with signs of sepsis, the CDC recommends a full diagnostic evaluation that includes the following22:
- Blood culture;
- Complete blood cell count (CBC) including white blood cell differential and platelet count;
- Lumbar puncture if the infant is stable enough to tolerate the procedure; and
- Chest x-ray if any respiratory problems present.
That said, debate remains regarding the value of such tools. For instance, CBC alone offers poor predictive value and is better for ruling out—through serial normal CBCs performed 8 to 12 hours apart, plus a negative blood culture at 24 hours—EOS.23
Additionally, neutrophil indices (absolute neutrophil count, immature to total neutrophil ratio, and absolute band) have proven more useful for ruling out rather than diagnosing infections.19 Because few conditions besides sepsis reduce neonatal neutrophil counts, neutropenia may be a better marker for neonatal sepsis than elevated neutrophil counts.24
Somewhat similarly, although bacteremic infants often have low platelet counts, such counts lack specificity and sensitivity, offering only a late marker of sepsis that doesn’t accurately gauge response to antimicrobial treatment.25 Also, acute-phase reactants such as C-reactive protein (CRP), cytokines, and procalcitonin generally offer high sensitivity but lack specificity to exclude sepsis when it is not present.26
The utility of urine cultures depends on timing. Infants with suspected EOS should not undergo urine cultures because positive cultures from neonates aged up to 6 days reflect bacteremia rather than an isolated urinary tract infection.19 Rather, urine culture obtained by catheter or bladder tap should be included in the sepsis evaluation for infants aged older than 6 days; in infants with suspected LOS, physicians also should obtain cultures from any other potential foci of infection.27
Asymptomatic infants
Asymptomatic neonates may require at least limited laboratory evaluations, depending on the presence or absence of certain risk factors (Table 2).19,27
Treatment options
When it comes to antimicrobial therapy for newborns with suspected sepsis, prospective, randomized, controlled trials remain relatively rare. Empirical antibiotic choices (Table 3) must rest on considerations including the specific organism associated with sepsis, as well as its sensitivities, and prevailing nosocomial infection trends in the nursery.16,19,22,28
Although third-generation cephalosporins (except ceftriaxone) represent a reasonable alternative to aminoglycosides, prolonged use creates a risk for invasive candidiasis.29 Routine use of the third-generation cephalosporin cefotaxime for EOS has been associated with rapid development of resistance.30 Accordingly, the AAP recommends limiting cefotaxime, which offers excellent cerebrospinal fluid penetration, to infants with gram-negative meningitis.
Duration of therapy
Because 98% of bacteremic cultures drawn before antibiotic therapy are positive by 72 hours, physicians typically continue empirical antimicrobial therapy for 48 to 72 hours pending culture results or development of clinical or hematologic signs of infection.28
When considering the duration of therapy in infants with negative blood cultures, says the AAP, physicians should factor in the infant’s clinical situation and risks associated with longer courses of antimicrobial therapy.
Supportive care
Finally, antibiotics are only part of the picture in managing neonatal sepsis. Physicians should never discount the importance of supportive care such as the following18:
- General: Thermal care, incubator, monitoring of oxygen saturation, heart rate, blood pressure, phototherapy if warranted.
- Respiratory: Support for apnea, hypoxia, respiratory distress.
- Cardiovascular: Plasma volume expanders (normal saline, 10-20 mL/kg initially); correction of fluid, electrolyte, glucose, and dermatologic imbalances (including blood, platelets, clotting factors).
- Gastrointestinal: Unstable infants generally need enteral feedings withheld.
Conclusion
Combating neonatal sepsis requires a high index of suspicion and, when indicated, timely appropriate treatment. This approach can help minimize the burden of neonatal sepsis on families, physicians, and the healthcare system alike.
